Statement on Monetary Policy – August 2020 Box A: Using Wage Subsidies to Support Labour Markets Through the COVID-19 Shock

Wage subsidy schemes have been a key part of the economic policy response to COVID-19

Global labour markets have been severely disrupted by the COVID-19 pandemic. In most countries, the public health response to the COVID-19 outbreak included the temporary closure of non-essential businesses to slow the spread of the virus, which directly reduced labour demand. Labour demand declined even further than this as many sectors that were still able to operate responded to weak demand for their output and disruptions from health concerns and supply chains.

As part of a policy package to support affected households and businesses, many advanced economies have turned to wage subsidy schemes. Wage subsidy programs have a long history in parts of Europe but have been used less extensively elsewhere. For example, Germany has had wage subsidy schemes since the 1920s and France has had one since the late 1960s. The nature of the COVID-19 shock, and the experience of some economies with wage subsidies (particularly Germany) during the global financial crisis (GFC) have both encouraged the use of these schemes more broadly. The Organisation for Economic Co-operation and Development (OECD) reported that as of June this year, 28 of its 37 member economies had used wage subsidies as the main support for workers facing reduced hours or the risk of being stood down because of COVID-19.

These programs typically involve the government subsidising wages to help employers retain employees during an economic shock, which preserves firm-specific human capital by maintaining the link between firms and their employees. This can speed up the recovery by reducing the need for firms to find new workers, who might be less suited to the available roles than the workers who were laid off.[1] The wage subsidy schemes increase firms' flexibility to respond to reduced demand and they can limit unemployment increases by spreading the loss of hours across a larger share of the labour force. The schemes also support employee incomes by topping up wages where employees are working reduced hours. International experience suggests that the long-term effects of the schemes can work in different directions: while keeping workers attached to their original employers could reduce the scarring effects from prolonged unemployment, it might also slow needed adjustments if the shock results in lasting structural changes that require significant labour force reallocation.

Following the initial COVID-19 outbreak, existing wage subsidy schemes were significantly enhanced. Eligibility was expanded to cover more sectors, a broader range of firms and more types of employment contracts. Some countries also introduced similar programs to support the self-employed. Many wage subsidy schemes allowed for a larger reduction in working hours than before the pandemic and were opened to less severely affected firms. Some schemes raised their payment caps and increased their replacement rates, which measure the ratio of income for employees in the scheme relative to their pre-scheme earnings. A few economies, including Australia, Canada, New Zealand and the United Kingdom, introduced wage subsidy schemes for the first time. Unlike most wage subsidy schemes that pay the covered employees a share of their pre-scheme earnings, the Australian and New Zealand schemes pay fixed amounts.

Wage subsidy schemes have supported a significant share of the labour force in advanced economies over recent months, far more than were supported during the GFC (Graph A.1).[2] The extent of support has varied across economies: around 60 per cent of the labour force in New Zealand, 15 to 45 per cent in the larger euro area economies, and 20 to 30 per in Canada and the United Kingdom. In Australia, more than a quarter of the labour force is covered by JobKeeper (for further discussion see the ‘Domestic Economic Conditions’ chapter). The United States is one of the few advanced economies that has not used a formal wage subsidy scheme, and instead has relied on significantly enhanced unemployment benefits to support household income.[3]

Graph A.1
Labour Market Impact

So far, standard measures of unemployment rates in advanced economies that have used wage subsidy schemes have remained around their pre-outbreak levels (Graph A.1). The unemployment rate in the euro area is little changed since February, while the average increase in OECD economies without wage subsidies has been around 3 percentage points and as high as 11 percentage points in the United States (although direct comparisons are hampered by measurement differences). Canada is the only economy with a comprehensive wage subsidy scheme that has also experienced a large increase in the unemployment rate. Some of the difference in the increase in unemployment results from an unrelated measurement difference, in that workers who have been temporarily stood down (who are not covered by the scheme) are immediately counted as unemployed in the US and Canadian data, but only with a lag elsewhere. But for Canada, the initial take-up of the scheme was reportedly slow due to its perceived complexity, and so the scheme was less effective in forestalling an initial wave of layoffs, temporary or otherwise. Historically, wage subsidy schemes have required participating employees to work some hours with their employer. However, the distinction between subsidised employment and unemployment has become less clear during the COVID-19 shock; due to the public health considerations, all these schemes have been opened to employees working zero hours.

Timely cross-country data on hours worked is limited, but it confirms that the main form of labour market adjustment in economies with wage subsidy schemes has been through a larger decline in average hours worked than usual, rather than a fall in employment.

It is too early to judge the medium-term success of wage subsidy schemes, but international organisations, such as the OECD and the International Monetary Fund, generally expect them to limit the increase in unemployment rates over the next two years.[4]

Wage subsidies have helped support household incomes and the nascent recovery in consumption as mandated containment measures have been unwound. The current wage subsidy schemes in Canada, the euro area and the United Kingdom generally compensate employees for a share of their gross pre-scheme earnings lost from the reduction in working hours. In most cases, the effective support to household income depends on actual hours worked and how it affects tax payable and other benefits given the complex interactions such schemes can have with the income tax system. Estimates for the euro area suggest that the after-tax replacement rates for employees in wage subsidy schemes working zero hours are between 60 to 80 per cent, which are comparable to the replacement rates from unemployment benefits in those countries (Graph A.2). However, it is likely that most participants in wage subsidy schemes will have more confidence they will return to their old job than an unemployed person has of getting a new job. Therefore, people receiving wage subsidies may be willing to spend more of their current income than an otherwise similar unemployed person.

Graph A.2
Net Wage Replacement Rates

Tapering the support

Governments have begun adjusting their wage subsidy schemes as their economies have moved beyond the acute phase of the COVID-19 shock, when lockdowns were the most stringent. Governments are faced with competing considerations when determining the duration of the wage subsidies. Prolonged use of wage subsidy schemes can have adverse long-run effects by reducing labour mobility and stymieing productivity growth.[5] These effects can be minimised by limiting the duration of the schemes; however, abrupt withdrawal of wage subsidies can lead to a sharp increase in unemployment and loss of incomes.

The duration of most schemes was set at around three months at the start of the mandated COVID-19 restrictions. The schemes have since been extended as it has become clearer that the recovery will be slow and uneven (Graph A.3). Some schemes have tightened eligibility at the same time as they have been extended. For instance, New Zealand requires firms to have experienced a bigger loss in revenues in the extended phase, while the United Kingdom has closed its schemes for new applicants and has introduced cost sharing with employers during the extended phase.

Graph A.3
Duration of Wage Subsidy Schemes


For more information on wage subsidies, see Lydon R, T Matha and S Millard (2019), ‘The whys and wherefores of short-time work: Evidence from 20 countries’, viewed 8 July 2020, available at <>. [1]

Comprehensive data on wage subsidy schemes is limited and often lagged. Most schemes publish timely data on the number of applicants to the scheme, which is used here to estimate how many workers were covered by the scheme at the height of its use; this may overestimate actual coverage if not all applicants access the scheme. [2]

The US Paycheck Protection Program (PPP) has been another key component of the US policy response and has some similarities to wage subsidy schemes but is structured very differently. The PPP extends forgivable loans to small businesses, via private sector banks. The forgiveness is achieved by maintaining employment at pre-outbreak levels but it is unclear how much employment PPP has already supported because borrowers have an extended period of time to qualify for forgiveness; estimates suggest that the PPP could support around 30 per cent of the US labour force. [3]

For further information, see OECD (2020), ‘Issue Note 5: Flattening the Unemployment Curve? Policies to Support Workers' Income and Promote a Speedy Labour Market Recovery’, Economic Outlook, 2020(1). Available at <> and IMF (2020), ‘April World Economic Outlook’. Available at <>. [4]

For a discussion of the potential adverse effects of wage subsidies, see A Arpaia, N Curci, E Meyermans, J Peschner and F Pierini (2010), ‘Short-time Working Arrangements as Response to Cyclical Fluctuations’, European Commission Occasional Paper, viewed 8 July 2020. Available at <>. [5]